The present invention relates to a method for identifying compounds which modulate sterol biosynthesis. In particular the invention provides in vivo assays for inhibitors of sterol biosynthesis wherein inhibition leads to a change in the level of expression of a reporter gene. The invention also relates to nucleic acids and recombinant cells for use in the above assays.
Two general approaches may be taken to the identification of inhibitors of a metabolic pathway. In the first approach, one or more individual enzymes from the pathway are selected, and compounds are screened for their ability to inhibit these enzymes in in vitro reactions. This approach can be labour intensive, and, depending on the assay and enzymes used, expensive. An alternative approach is to construct a system which reports on the activity of the entire metabolic pathway, either by directly measuring the levels of the product of the pathway, or by using an indirect measurement of these levels. In both prokaryotes and eukaryotes the products of a metabolic pathway sometimes regulate the expression of the genes which encode the enzymes in the pathway. By coupling the regulatory regions of these genes to reporter genes it is possible to obtain reporter gene assays for inhibitors of metabolic pathways.
In mammalian cells, the product of sterol biosynthesis (cholesterol) regulates the transcription of the genes encoding 3-hydroxy-3-methylglutaryl coenzyme A (HMG-CoA) synthase and HMG-CoA reductase. The sequences within the promoter regions of these genes which mediate the regulation of transcription in response to sterols have been determined, and are referred to as sterol response elements (SREs: Brown and Goldstein (1990) Nature 343, 425-430). A transcription factor which binds to these sequences has been characterised, and a mechanism for sterol regulation of the activity of this factor, involving sterol-regulated proteolysis, has been proposed (Wang et al (1994) Cell 77, 53-62). Construction of sterol-responsive reporter genes using SRE sequences has been described. For example, in mammalian cells, additional of exogenous sterols inhibits sterol biosynthesis and will also inhibit the activity of SRE-driven reporter genes (Osborne et al (1985) Cell 42, 203-212).
The regulation of genes encoding enzymes in fungal sterol biosynthesis is less well characterised. No sequence element equivalent to the SRE has been defined. HMG-CoA reductase is subject to feedback regulation by sterol levels in the budding yeast Saccharomyces cerevisiae (S. cerevisiae), but this regulation occurs at the translational level and is mediated by the 5xe2x80x2 untranslated region of the HMG1 mRNA (Dimster-Denk et al (1994) Mol. Biol. Cell 5, 655-665). Limited studies have been performed which examine the activities of other enzymes in the sterol biosynthetic pathway in the S. cerevisiae under conditions which inhibit ergosterol production. Squalene synthase levels are increased 2-3 fold by application of the HMG-CoA reductase inhibitor lovastatin (Robinson et al (1993) Mol. Cell Biol. 13, 2706-2717), and squalene epoxidase activity can increase up to 5 fold in mutant yeast strains containing limited sterol amounts. Messenger RNA levels of lanosterol 14-a-demethylase accumulate during anaerobic growth (Turi and Loper (1992) J. Biol. Chem. 267, 2046-2052); induction of this gene forms the basis for a patent application by Kirsch (EP 0627 491 A1) for a screen for inhibitors of ergosterol biosynthesis. This screen, which also requires additional manipulation of the yeast to reduce flux through the sterol biosynthetic pathway (by deletion of the HMG1 gene), only appears to detect compounds acting on lanosterol 14-a-demethylase or further down the sterol biosynthesis pathway. For example, tolnaftate, an inhibitor of squalene monooxygenase, was not detected in this screen (Kirsch, p. cit.).
We have now devised a assay which is capable of detecting a wide range of inhibitors of sterol biosynthesis. In this assay inhibition of ergosterol synthesis results in the induction of reporter gene activity in a yeast tester strain. The assay is simple, cheap and robust and may be employed in high throughput mode to screen large chemical collections, natural product collections and compound libraries. Sterol biosynthesis inhibitors have a number of applications. As inhibitors of an essential process in eukaryotes they are useful as antifungal drugs and agrochemicals, and may also have applications as biocides and as anti-parasitic agents. In humans, cholesterol may be synthesised de novo or absorbed from the intestine. Inhibition of de novo synthesis reduces lipid levels in the bloodstream, resulting in reduced risk, of atherosclerosis. Sterol biosynthesis provides metabolic intermediates which are used in lipid modification (farnesylation and geranylgeranylation) of proteins. Inhibition of the early stages of sterol biosynthesis may indirectly inhibit lipid modification. This may be of therapeutic application, for example in inhibiting tumour growth driven by activated ras genes.
Therefore in a first aspect of the present invention we provide a method for the identification of agents which modulate sterol biosynthesis which method comprises contacting a test compound with a host cell comprising a DNA sequence which controls expression of a yeast acetoacetyl CoA thiolase (ACoAT) gene operably linked to a reporter system such that modulation of sterol biosythesis in the host cell leads to a detectable change in cell phenotype, and determining whether any such detectable change has occurred.
By xe2x80x9coperably linkedxe2x80x9d we mean linked in such a way as to provide the basic sequence signals necessary for initiation of gene transcription and initiation of gene translation.
By xe2x80x9ca DNA sequence which controls expressionxe2x80x9d we mean a sequence which confers responsiveness of the activity of adjacent genes to change(s) in intracellular metabolic pathways such as sterol biosynthesis. In general such sequences are protein binding sites. Such proteins may regulate events such as transcriptional activation. Their ability to perform this regulation may be influenced by intracellular processes which feedback from sterol biosynthesis.
Acetoacetyl-CoA thiolase (EC 2.3.1.9) catalyses the condensation of two acetyl-CoA molecules to acetoacetyl-CoA. This enzyme precedes HMG-CoA synthase in the sterol biosynthetic pathway. There have been no studies of the regulation of the expression of this gene, and only limited studies on the regulation of the enzyme activity. ACoAT activity can be repressed 12 fold by addition of exogenous sterol (Trocha and Sprinson, Arch. Biochem. Biophys. (1976) 174, 45-51). Servouse and Karst (Biochem. J. (1986) 240, 541-547) provide similar evidence, showing that ergosterol starvation increases ACoAT activity, whereas ergosterol excess reduces ACoAT activity. These studies examined enzyme activities rather than the expression of the ACoAT gene, and thus it is not possible to determine whether the regulation is due to alterations in the enzyme itself, such as covalent modifications, or the binding of a modulator molecule, or whether the level of the protein is regulated through control of gene expression. The ACoAT gene was cloned from the brewing yeast Saccharomyces uvarum (Dequin et al (1988) Curr. Genet. 13, 471-478). These authors report that this gene is highly expressed. Furthermore, overexpression of this gene did not lead to an increase in ergosterol production, leading Dequin et al to conclude that ACoAT is not a rate-limiting enzyme in the production of ergosterol. Despite the availability of DNA probes which could be used to measure mRNA levels under different conditions, no studies of the expression of the ACoAT gene in fungi have been conducted under conditions of ergosterol depletion through anaerobic, mutant or inhibitor conditions. We have found that when the promoter region of S. cerevisiae acetoacetyl CoA thiolase is linked to a reporter gene the reporter gene may be induced by sterol biosynthesis inhibitors. This assay provides a convenient, cheap and robust screen for novel inhibitors of sterol biosynthesis.
Fungal species which may be useful in either the provision of an acetoacetyl-CoA thiolase promoter, or which may provide a host cell suitable for use in the construction of a reporter gene assay, include Saccharomyces species such as Saccharomyces cerevisiae and Saccharomyces uvarum, Schizosaccharomyces pombe, Candida species, Cryptococcus neoformans, Histoplasma capsulatum, Pneumocystis carinii, Neurospora crassa, Septoria species, Magnaporthe grisea, Aspergillus species, Ustilago species and Botrylis cinerea. 
The assay may be potentially used to detect inhibitors of any of the enzymes involved in fungal sterol biosynthesis. These enzymes include acetoacetylCoA thiolase, HMG-CoA synthase, HMG-CoA reductase, mevalonate kinase, mevalonate-5-phosphate kinase, mevalonate-5-phosphate decarboxylase, squalene synthase, squalene epoxidase, 2,3-oxidosqualene cyclase, C24-methyltransferase, C14-demethylase, D14-reductase, D8-7 isomerase, C5-desaturase and C24(28)reductase. Although the assay is established in fungal cells, inhibitors of mammalian enzymes involved in sterol biosynthesis may be identified either through the ability of compounds to inhibit both the fungal and mammalian homologues of a particular enzyme, or through use of fungal cells engineered to express a mammalian enzyme from the sterol biosynthesis pathway.
In a preferred aspect of the present invention inhibitors of sterol biosynthesis lead to reporter system activation. The advantage of using a reporter gene as a reporter system is that it confers a readily measurable phenotype upon the cell. The reporter gene may conveniently comprise the coding sequence of a enzyme such as E. coli xcex2-galactosidase (Casadaban et al., (1983) Meth. Enzymol. 100, 293-308), firefly luciferase (de Wet et al.,(1987) Mol. Cell Biol. 7, 725-737), or E. coli choramphenicol acetyl transferase (Gorman et al., (1982) Mol. Cell Biol. 2, 1044-1051), or green fluorescent protein, in which the phenotype conferred may be measured by alterations in fluorescence (Chalfie et al., (1994) Science 263, 802-805).
Alternatively the reporter system may comprise a gene essential for the growth of the cell, for example a gene encoding an essential metabolic enzyme. In this case, activation of the gene will allow cell growth. Alternatively the reporter gene may encode an enzyme which metabolises a toxic substrate. In the presence of this substrate, activation of the reporter gene will again allow growth of the cell. Therefore in a further aspect of the invention we provide an assay for agents which modulate sterol biosynthesis in which the output is the growth of the reporter cell.
The assay described here may be used in combination with another reporter system in the same cell, allowing for compounds to be simultaneously screened for the ability to modulate sterol biosynthesis and other processes. The recent disclosure of novel forms of green fluorescent proteins with different absoption spectra allows the use of multifunctional assays in the cell using the same output (Delagrave et al (1995) Bio/Technology 13, 151-154).
The reporter gene assays may be performed in a variety of ways. In Example 1 we describe the use of liquid reporter gene assays in microtitre plates. Assays may also be performed on solid media in indicator plates, in which positive compounds may appear as zones of colour, the size of the zone being a measure of the potency of the compound. Such zone screening approaches may be particularly applicable to the use of the reporter assay in screening combinatorial libraries or compounds synthesised on solid media (Ecker and Crooke (1995) Bio/Technology 13, 351-360).
In a further aspect of the invention we provide reporter cells containing the ACoAT-based reporter gene which may also contain mutations or other additional genes which alter the rate of flux through the sterol biosynthesis pathway and hence alter the sensitivity of the assay to sterol biosynthesis inhibitors. We also provide cells which contain mutations, such as cell wall mutations, which increase the permeability of the fungal cell wall to compounds and other exogenous agents. Such mutations may be selected by mutagenising a reporter strain such as that described in Example 1, and selecting cells which show an enhanced reponse to sterol biosynthesis inhibitors.
In a further aspect of the invention we provide cells which contain the reporter system, and which also contain genes encoding enzymes of the sterol biosynthesis pathway from other organisms. It may be desirable to express such genes in combination with disruptions and/or mutations of host genes. For example, the host gene encoding a certain enzyme may be disrupted and a gene encoding the homologous human enzyme introduced. Compounds which are active against the xe2x80x9chumanisedxe2x80x9d reporter strain, but inactive against the parental strain containing the fungal homologue of the enzyme, are likely to be acting by inhibiting the human form of this enzyme.
In the example below we describe the construction of an S. cerevisiae strain in which the reporter gene is integrated into the host genome. Various alternative methods of maintaining reporter genes in fungal cells include the use of multicopy and single copy episomal plasmids.
In the reporter strain described in Example 1, in which the promoter of an acetoactyl CoA thiolase gene is linked to a reporter gene, inhibition of sterol biosynthesis results in stimulation of reporter gene output. This reporter gene output allows further definition of the promoter elements which mediate the response to reductions in sterol levels. There are at least two ways in which such information could lead to the construction of more sensitive screens for sterol biosynthesis inhibitors:
(1) yeast promoter regions can be quite complex, containing binding sites for multiple proteins, some of which may act as activators of gene expression, and some of which may act as repressors of gene expression (reviewed by Struhl (1989) Ann. Rev. Biochem. 58, 1051-1077 and Curr. Opin. Cell Biol. (1993) 5, 513-520). The definition of activator and repressor regions by construction of deletion mutants would allow the construction of synthetic promoters in which the repressor regions had been removed. This could enhance reporter gene output in response to sterol biosynthesis inhibition and may increase the sensitivity of the screen.
(2) the definition of an element which mediates the effect of sterol biosynthesis inhibition on reporter gene activation could lead to the development of new improved promoters based on this element. For example, the S. cerevisiae GAL1 gene was found to inducible by galactose (St. John and Davis (1981) J. Mol. Biol. 152, 285-315). The galactose responsive region was mapped by deletion analysis (West et al (1984) Mol. Cell Biol. 4, 2467-2478). Within this region, four homologous sequences were found to be binding sites for the transcription factor GAL4 (Giniger et al (1985) Cell 40, 767-774). Comparison of these sequences led to the deduction of a consensus binding site for GAL4, and has allowed the synthesis of related sites which do not occur naturally but which are strong binding sites for GALA (Marmorstein et al (1992) Nature 356, 408-414). With this knowledge of the GAL4-DNA interaction it has been possible to construct synthetic promoters which are highly responsive to GALA by either multimerising the number of GALA binding sites, or using strong binding sites. In a similar way, elements in the ACoAT promoter which mediate the effect of sterol biosynthesis inhibition may be binding sites for transcription factors, and systematic mutagenesis of such sequences may lead to the definition of a consensus response element, the multimerisation of which may be used to construct reporter strains which are more sensitive to sterol biosynthesis inhibition. Therefore in a further aspect of the invention we provide reporter cells in which the reporter system is sensitive to sterol biosynthesis inhibition, and in which the promoter for the reporter system contains one or more sequence elements from an ACoAT promoter, or one or more sequence elements derived from analysis of the ACoAT promoter.
Identification of the sterol response element that mediates the effect will allow the cloning of proteins which bind this element. Cloning of such DNA binding proteins may be performed by a variety of methods. cDNA expression libraries may be screened with the appropriate DNA fragment (Singh et al (1988) Cell 52, 415-423). Alternatively, the reporter strain may be subjected to mutagenesis and mutants isolated which lack the reporter gene response. Cloning of genes which complement these mutaions may identify DNA binding proteins (Rose and Broach pp 195-230 of Guide to yeast genetics and moleculr biology (Guthrie and Fink eds.) Meth. Enzymol. vol. 194 (1991) Academic Press). Also, the xe2x80x9cone-hybridxe2x80x9d system, involving fusion of cDNA libraries to a transcriptional activator region, may be employed (Wang and Reed (1993) Nature 364, 121-126). Cloning of factors which mediate the sterol responsiveness of the ACoAT gene may lead to the characterisation of the mechanism of feedback regulation of sterol biosynthesis in fungi, and may also elucidate new targets for drug action.
The invention will now be illustrated but not limited by reference to the following detailed description, Example, and Figures:
The S. cerevisiae ACoAT gene (ERG 10) has been cloned (Hiser et al (1994) J. Biol. Chem. 269, 31383-31389) and the DNA sequence deposited in the Genbank-EMBL database (Hiser et al Accession no. L20428, 30th Jun. 1994). Hiser et al disclose approximately 500 bp of sequence upstream of the start site of translation of the ACoAT gene. We refer to this region as the 5xe2x80x2 flanking region of the ACoAT gene. We have cloned the 5xe2x80x2 flanking region region and fused it to a reporter gene (Escherishia coli (E. coli) b-galactosidase). This gene was introduced into S. cerevisiae cells to make a reporter yeast strain. In the reporter strain we find that the activity of the reporter gene when grown under aerobic conditions in the absence of inhibitors of sterol biosynthesis is low. This contrasts with the report of Dequin et al (op. cit.) that the ACoAT gene is highly expressed in S. uvarum. After incubation with sterol biosynthesis inhibitors for a period of several hours, reporter gene activity is induced considerably. Up to 9-fold induction can be obtained in this way. Such induction occurs with several compounds which are known to inhibit sterol biosynthesis, including those which act early in sterol biosynthesis (before the sterol nucleus has been formed), and those which act at later steps (after the sterol nucleus has been formed). The induction of reporter gene activity in this yeast reporter strain is useful for screening compound collections for novel inhibitors of sterol biosynthesis. These results also provide evidence that ACoAT activity is feedback regulated by sterol levels at least in part through regulation of gene transcription, and that at least some of the DNA elements which mediate this effect reside within the promoter region used to construct the reporter gene.
The reporter gene was constructed using 5xe2x80x2 flanking sequence disclosed in the deposition of Hiser et al. In the deposited sequence of S. cerevisiae ACoAT gene, the coding region starts at position 548.The reporter gene described here was constructed from a DNA fragment containing nucleotides 7-543 of this sequence. Several alternative fragments of the gene could have been chosen for construction of the reporter gene. Eukaryotic promoters consist of a number of binding sites for trans-acting factors. The binding of these factors determines both the site at which transcription initiates and the level of transcription from that site. The site of transcription initiation in the ACoAT gene was not mapped by Dequin et al (op. cit.) However, a consensus sequence for a TATA box, TATAAA, is present at position 419-424 and is also present at the homologous position in the S. cerevisiae gene. The TATAAA sequence, a binding site for the general transcription factor TFIID, positions the start site of transcription (Nagawa and Fink (1985) Proc. Natl. Acad. Sci. USA 82, 8557-8561; Hahn et al (1985) Proc. Natl. Acad. Sci. USA 82, 8562-8566). Transcription usually initiates betwenn 60 and 120 bp downstream of the TATA box in S. cerevisiae. It appears most probable that transcription initiates at around position 490, since this is the location of an xe2x80x9cRRYRRxe2x80x9d sequence, used as the transcription start site in 45% of S. cerevisiae genes analysed (Hahn et al, op. cit.). Thus the fragment of 5xe2x80x2 flanking region employed in the construction of the reporter gene in Example 1 was chosen because of the likelihood that it would confer efficient transcriptional initiation together with any regulation by sterol levels. A number of alternative ways of using the 5xe2x80x2 flanking region of the ACoAT gene to construct the reporter gene can be envisaged. Fusion points downstream of the translation initiation codon of ACoAT could also be used, but would need to be chosen with care so that the encoded fragment of the thiolase gene is in frame with the reporter gene coding region. Fusion points upstream of that used in Example 1 may also be functional, though if they lack the TATAAA box or regulatory sequences, they may not function as well as the reporter gene described in Example 1. A possible binding site for a transcription factor is located at positions 373-382. This sequence, SEQ ID NO:1 CGTGGCCAGG, is an imperfect inverted repeat which is conserved in the S. uvarum promoter region. Inverted repeat structures are potential binding sites for dimeric DNA binding proteins. We also note that some 5xe2x80x2 untranslated region of the ACoAT gene is included in the reporter construct, and that effects of sterol biosynthesis inhibitors on reporter gene expression could, in theory, be due effects on mRNA translation mediated by this sequence at the RNA level, rather than by effects on transcription.
The cis-acting sequences which mediate the induction of reporter gene expression in response to sterol biosynthesis could be defined initially by creating deletions of the ACoAT promoter and fusing them to a reporter gene. Some information about potential protein binding sites in the promoter region may be obtained by analysing the sequence for palidromic sequences (which may be binding sites for dimeric DNA binding proteins) and sequences conserved between the promoter regions of other genes. The sequence SEQ ID NO:1 CGTGGCCAGG (positions 372-381 of FIG. 11 SEQ ID NO:4, positions 373-382 of Hiser et al Genbank-EMBL Accession no. L20428)) is a candidate for a binding site for a dimeric DNA binding protein, by virtue of its near-palindromic nature. This sequence is conserved in the promoter of the ACoAT gene from S. uvarum. Furthermore, an 80% match to this sequence is also found in the promoter region of the S. cerevisiae CYC1 gene (positions 171-180 of Genbank Accession no. M11345). In a yeast strain in which the ERG3 (C5 desaturase) gene is deleted, the CYC1 gene is induced 6 fold (Parks and Smith (1995) Yeast 11, S311). In this strain ergosterol biosynthesis is inhibited by the ERG3 gene deletion and this results in induction of the CYC1 gene. It is possible that the induction of CYC1 in the ERG3-deleted strain and reporter gene induction in the reporter strain MEY133::pACoAT by inhibition of sterol biosynthesis occurs by similar mechanisms, in which case the sequence element identified above could play a role. The CYC1 and ACoAT gene share only one other short region of homology within 150 bp upstream of their putative TATA boxes. This region, which is not palindromic, occurs at positions 387-403 of ACoAT (128-144 of CYC1). However, neither this sequence, nor the near palindrome SEQ ID NO:1 CGTGGCCAGG discussed above, are found within the upstream regions of other genes encoding enzymes in the sterol biosynthesis pathway, such as ERG3 (C5 desaturase), ERG1 (squalene monooxygenase), ERG8 (phosphomevalonate kinase) and ERG9 (squalene synthase). Thus, an analysis of promoter regions of genes encoding enzymes of the sterol biosynthesis pathway does not reveal a common sequence element which is likely to be the equivalent of the mammalian sterol response element. It is difficult to predict which regions of the fragment used in construction of the reporter plasmid pACoAT are necessary for the reporter gene response to sterol biosynthesis inhibition. It is likely that the putative TATA box is necessary for accurate transcription initiation. If there is any translational regulation of gene expression, then the 5xe2x80x2 noncoding region of the ACoAT mRNA (positions 490-543) is likely to be important. Regulation at the level of transcription initiation would by contrast most likely involve sequences upstream of the TATA box (xe2x80x9cupstream activating sequencesxe2x80x9d; Struhl, op. cit.).
In Example 1 we describe the construction of an S. cerevisiae reporter strain in which the reporter gene is integrated at the ura3 locus. Various alternative modes of constructing and introducing reporter genes into S. cerevisiae will be apparent to the person of ordinary skill. The reporter gene should ideally consist of the 5xe2x80x2 flanking region operationally linked to the coding region of a reporter gene. Downstream of the coding region it may be desirable to have a terminator sequence from an S. cerevisiae gene. The reporter gene may be introduced by integration into the host genome, or it may be located on a single or multicopy plasmid. A variety of common laboratory strains of S. cerevisiae may be used to establish the screen, provided that they contain host mutations which allow the introduction and maintenance of foreign DNA. Introduction of mutations in genes encoding enzymes in the sterol biosynthetic pathway may alter the flux through the pathway and alter the signal to noise ratio of the reporter gene in the assay. The introduction of mutations in genes involved in cell wall biosynthesis, or the mating of the reporter strain with strains displaying an enhanced permeability to compounds, may increase the permeability of the reporter strain to compounds which can inhibit sterol biosynthesis and activate the reporter gene. The reporter strain itself may be made more sensitive to compounds by mutagenising the strain and selecting mutants which show enhanced reporter gene output to positive compounds.
Example 1 shows a summary of the compounds used in FIGS. 2-10, the enzymes which they inhibit, the overall level of reporter gene induction at a concentration of 100 xcexcg/ml, and the reporter gene induction per cell at a concentration of 100 xcexcg/ml. Reporter gene induction is defined as:
[OD570f100xe2x88x92OD570i100]/[OD570f10xe2x88x92OD570i10]
where f refers to a final reading, and i to an initial reading. 100 to a concentration of 100 xcexcg/ml and 10 to a concentration of 10 ng/ml. Reporter gene induction per cell is defined as:
[{OD570f100xe2x88x92OD570i100}/OD570i100]/[{OD570f10xe2x88x92OD570i100}/OD570i10]
FIG. 1 summarises the structure of a reporter gene containing 5xe2x80x2 flanking sequences from the ACoAT gene. Nucleotides 7-543 of the S. cerevisiae ACoAT gene were obtained as a PCR fragment with a SalI site at the 5xe2x80x2 end and a HindIII site at the 3xe2x80x2 end. This was inserted between the SalI and HindIII sites of the plasmid pJP 159 to create the plasmid pACoAT. TATA indicates the position of a TATA sequence, a likely binding site for TATA binding protein. The coding region of xcex2-galactosidase is indicated by a shaded region, with ATG marking an initiator methionine. There are 112 bp between the HindIII site and the methionine. The sequence in this region derives from the plasmid pRY171. The numbers derive from the sequence of ACoAT.
FIG. 2 shows dose-responses of reporter gene activity and growth for lovastatin. Reporter gene activity (OD570finalxe2x88x92OD570initial) is indicated by solid squares. Growth (OD570 final) is indicated by open squares. Each point is the average of three measurements. Error bars indicate standard deviations.
FIG. 3 shows dose-responses of reporter gene activity and growth for terbinafine. Nomenclature is as in FIG. 2.
FIG. 4 shows dose-responses of reporter gene activity and growth for squalestatin. Nomenclature is as in FIG. 2.
FIG. 5 shows dose-responses of reporter gene activity and growth for 4-t-butyl-N-methyl-N-(1-naphthylmethyl)benzylamine. Nomenclature is as in FIG. 2.
FIG. 6 shows dose-responses of reporter gene activity and growth for flutriafol. Nomenclature is as in FIG. 2.
FIG. 7 shows dose-responses of reporter gene activity and growth for amorolfine. Nomenclature is as in FIG. 2.
FIG. 8 shows dose-responses of reporter gene activity and growth for fenpropimorph. Nomenclature is as in FIG. 2.
FIG. 9 shows dose-responses of reporter gene activity and growth for 3-(4-t-butylbenzyl)-(N-hexyl)-piperidine. Nomenclature is as in FIG. 2.
FIG. 10 shows dose-responses of reporter gene activity and growth for 3-(4-t-butylphenyl)-7-isopropyl-N-methylisoquinoline. Nomenclature is as in FIG. 2.
FIG. 11 shows the structure of plasmid pACoAT as referred to in the legend to FIG. 1 and in FIG. 1. The structure of this 7.6 kb plasmid is shown to scale, together with selected restriction enzyme cleavage sites. The open reading frame of E. coli b-galactosidase is located between the HindIII and KpnI sites and is denoted by the arrow marked LacZ(ORF). The S. cerevisiae URA3 gene is located between the KpnI and BglII sites and contains a unique ApaI site used to linearise the DNA for transformation of yeast. Between BglII and XhoI is the backbone of the plasmid, derived from SP72 and containing the Ampicillin resistance gene. Between XhoI and SalI is a terminator from the GAL11 gene. Finally, the ACoAT promoter, as shown in SEQ ID NO:4, is located between the SalI and HindIII sites and is marked by the arrow labelled xe2x80x9cACoAxe2x80x9d.